Artificial intelligence-based recyclable material treatment apparatus and control method thereof

Abstract

The present disclosure relates to an artificial intelligence-based recyclable material treatment apparatus and a control method thereof, and can automatically determine whether a recyclable material satisfies deposit refund criteria on the basis of a captured image of the recyclable material. Project Unique Number (10 digits): 2710071113 Project Number: RS-2024-00460108 Ministry: Ministry of Science and ICT Project Management (Specialized) Agency: Institute of Information & Communications Technology Planning & Evaluation (IITP) Research Program Name: SW Computing Industry Core Technology Development (R&D, Informatization) Research Project Title: Development of an Artificial Intelligence-based National / Regional Customized Recycling Education and Automatic Recycling Waste Classification System Performing Organization: Seoreu Co., Ltd. Research Period: July 01, 2024 – December 31, 2025 (1 year and 6 months)

Description

Artificial intelligence-based recycling processing device and control method thereof

[0001] The present disclosure relates to an artificial intelligence-based recycling processing device and a method for controlling the same.

[0002] As the importance of recycling processing systems for environmental protection and resource circulation increases, various automated sorting and recovery devices are being developed. In particular, technologies utilizing image processing and artificial intelligence (AI) to automatically classify recyclables with specific shapes, such as cans, glass bottles, and plastic containers, and to improve recovery rates are being commercialized.

[0003] Meanwhile, some countries and regions operate a container deposit refund system in which a deposit is charged upon product purchase and refunded upon the return of the recyclable item after use. To operate such a system efficiently, it is essential to go beyond simple item classification and accurately determine whether the recyclable item is eligible for a deposit refund.

[0004] However, most existing recycling processing systems are limited to classifying items based on basic external appearance, such as cans, bottles, and plastics, and fail to extend to determining deposit refund conditions. In particular, since deposit refund conditions are applied differently depending on region and product characteristics, as described below, there is a problem in that accurate determination is difficult based solely on simple shape information.

[0005] For example, even for cans of the same shape, recycling refund conditions may differ in different countries or regions, and there are cases where certain product groups are eligible for refund only if their capacity exceeds a certain standard.

[0006] To address these issues, there is a growing need for advanced technology that utilizes artificial intelligence to classify recyclable items and comprehensively determine eligibility for deposit refunds.

[0007] The embodiments disclosed in this disclosure are intended to provide an artificial intelligence-based recycling processing device.

[0008] In addition, the embodiments disclosed in this disclosure aim to provide a recycling processing device capable of automatically determining whether a recyclable item meets the deposit refund criteria based on a captured image of the recyclable item.

[0009] The problems that this disclosure aims to solve are not limited to those mentioned above, and other unmentioned problems will be clearly understood by a person skilled in the art from the description below.

[0010] A recycling processing device according to one embodiment of the present disclosure for solving the above-described problem comprises: a conveying unit for conveying a recycling item; at least one camera for photographing a recycling item conveyed by the conveying unit; and a processor that determines the type of the recycling item based on an image captured by the camera using an artificial intelligence model, determines one or more attributes related to the deposit refund criteria of the recycling item based on the determined type, at least one shape information and at least one identification information regarding the recycling item obtainable through the image, and determines whether the determined attribute satisfies the deposit refund condition.

[0011] In addition, the identification information includes at least one of a label, character, barcode, logo, or color information included in the recycled item.

[0012] In addition, the processor can determine the attributes necessary for determining the deposit refund conditions by obtaining first shape information of the recyclable item through the image information and inferring complementary second shape information based on the first shape information by referring to standard shape information corresponding to the type of recyclable item determined above.

[0013] In addition, the shape information includes at least one of the outer contour, aspect ratio, height, width, or whether the recycled product is curved.

[0014] In addition, the processor may calculate at least one of the capacity or weight of the recycled material from the shape information and determine whether the deposit refund condition is satisfied based on at least one of the calculated capacity and weight.

[0015] In addition, the processor can determine whether a product is eligible for regional refund based on at least one of a label, character, logo, or color information included on the surface of the recycled product.

[0016] In addition, the above-mentioned conveying unit includes a conveyor partitioned at regular intervals, and the processor can use the partition structure as a reference value for determining the shape information.

[0017] In addition, the above deposit refund criteria include whether the product belongs to a product group for which a deposit is imposed at the time of sale in the relevant region or country.

[0018] In addition, the above attribute is a value related to the capacity or weight of the recyclable material, and the processor can determine whether the refund condition is met based on whether the attribute satisfies the value defined in the deposit refund criteria.

[0019] In addition, the processor controls a camera to acquire a plurality of images for the recognition of the recyclable material, inputs the plurality of images into a recognition model to aggregate the calculated recognition results, determines the final recognition result by a majority algorithm, and can store at least one of the images for which a recognition result matching the determined final recognition result is produced as a log image.

[0020] In addition, the processor can store the image with the highest recognition accuracy among the plurality of images as the log image.

[0021] In addition, the processor can aggregate the cumulative number of recognized recyclables by type and display the aggregated cumulative number through a user interface.

[0022] In addition, the processor stores a recognition image of a recyclable item obtained through the camera, and when a user selects a specific recyclable item, it can provide an image or video of the section where the selected recyclable item appears.

[0023] In addition, a control method for a recycling processing device according to one embodiment of the present disclosure for solving the above-described problem comprises: a step of controlling a conveying unit to convey a recycling item; a step of controlling at least one camera to photograph the recycling item conveyed by the conveying unit; a step of determining the type of the recycling item based on an image captured through the camera using an artificial intelligence model; a step of determining an attribute related to the deposit refund criteria of the recycling item based on at least one shape information of the recycling item obtainable through the determined type and the image; and a step of determining whether the determined attribute satisfies the deposit refund condition.

[0024] In addition, a computer program stored on a computer-readable recording medium for executing a method for implementing the present disclosure may be further provided.

[0025] In addition, a computer-readable recording medium for recording a computer program for executing a method for implementing the present disclosure may be further provided.

[0026] According to the aforementioned means for solving the problem of the present disclosure, an artificial intelligence-based recycling processing device can be provided.

[0027] In addition, according to the aforementioned means for solving the problem of the present disclosure, it is possible to automatically determine whether a recyclable item meets the deposit refund criteria based on a captured image of the recyclable item.

[0028] The effects of the present disclosure are not limited to those mentioned above, and other unmentioned effects will be clearly understood by a person skilled in the art from the description below.

[0029] FIG. 1 is a drawing illustrating the automatic refund of a deposit according to a recycling processing device according to an embodiment of the present disclosure.

[0030] FIG. 2 is a block diagram of a recycling processing device according to an embodiment of the present disclosure.

[0031] FIG. 3 is a flowchart of a recycling processing method according to an embodiment of the present disclosure.

[0032] FIGS. 4 to 6 are drawings illustrating the compression of recycled materials transported to a compression device by a transport unit.

[0033] FIG. 7 is a drawing illustrating an image of a recycled item being transported by a transport unit (150).

[0034] Figure 8 is a diagram illustrating the final recognition results and refund details for processed recyclables.

[0035] Figure 9 is a diagram illustrating a record of processing through a recycling processing device.

[0036] Throughout this disclosure, the same reference numerals denote the same components. This disclosure does not describe all elements of the embodiments, and general content in the art to which this disclosure pertains or content that overlaps between embodiments is omitted. The terms 'part, module, component, block' as used in the specification may be implemented in software or hardware, and depending on the embodiments, a plurality of 'parts, modules, components, blocks' may be implemented as a single component, or a single 'part, module, component, block' may include a plurality of components.

[0037] Throughout the specification, when a part is described as being "connected" to another part, this includes not only cases where they are directly connected but also cases where they are indirectly connected, and indirect connections include connections made via a wireless communication network.

[0038] Furthermore, when it is stated that a part "includes" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components.

[0039] Throughout the specification, when it is stated that a component is located "on" another component, this includes not only cases where a component is in contact with another component, but also cases where another component exists between the two components.

[0040] Terms such as "first," "second," etc., are used to distinguish one component from another, and the components are not limited by the aforementioned terms.

[0041] Singular expressions include plural expressions unless there is an obvious exception in the context.

[0042] In each step, identification codes are used for convenience of explanation and do not describe the order of the steps; the steps may be performed differently from the specified order unless a specific order is clearly indicated in the context.

[0043] The operating principles and embodiments of the present disclosure will be described below with reference to the attached drawings.

[0044] In this specification, the 'recyclable material processing device according to the present disclosure' includes all various devices capable of performing computational processing and providing results to a user. For example, the recyclable material processing device according to the present disclosure may include a computer, a server device, and a portable terminal, or may take the form of any one of them.

[0045] Here, the computer may include, for example, a laptop, desktop, notebook, tablet PC, slate PC, etc. equipped with a web browser.

[0046] A server device is a server that processes information by communicating with external devices, and may include an application server, a computing server, a database server, a file server, a game server, a mail server, a proxy server, and a web server.

[0047] A portable terminal is a wireless communication device that ensures portability and mobility, and may include all kinds of handheld-based wireless communication devices such as PCS, GSM, PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), WiBro (Wireless Broadband Internet) terminals, smartphones, etc., as well as wearable devices such as watches, rings, bracelets, anklets, necklaces, glasses, contact lenses, or head-mounted devices (HMDs).

[0048] The artificial intelligence-related functions according to the present disclosure are operated through a processor and a storage unit. The processor may be composed of one or more processors. In this case, the one or more processors may be general-purpose processors such as CPUs, APs, and DSPs (Digital Signal Processors), graphics-dedicated processors such as GPUs and VPUs (Vision Processing Units), or artificial intelligence-dedicated processors such as NPUs. The one or more processors control the processing of input data according to predefined operation rules or artificial intelligence models stored in the storage unit. Alternatively, if the one or more processors are artificial intelligence-dedicated processors, the artificial intelligence-dedicated processors may be designed with a hardware structure specialized for processing a specific artificial intelligence model.

[0049] The predefined rules of operation or artificial intelligence models are characterized by being created through learning. Here, being created through learning means that a basic artificial intelligence model is trained using a number of training data by a learning algorithm, thereby creating predefined rules of operation or artificial intelligence models configured to perform desired characteristics (or objectives). Such learning may be performed on the device itself where the artificial intelligence according to the present disclosure is executed, or it may be performed through a separate server and / or system. Examples of learning algorithms include supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning, but are not limited to the examples described above.

[0050] An artificial intelligence model may be composed of multiple neural network layers. Each of the multiple neural network layers has multiple weights and performs neural network operations through operations between the results of previous layers and the multiple weights. The multiple weights possessed by the multiple neural network layers can be optimized based on the learning results of the artificial intelligence model. For example, the multiple weights can be updated so that the loss value or cost value obtained by the artificial intelligence model is reduced or minimized during the learning process. Artificial neural networks may include deep neural networks (DNNs), such as Convolutional Neural Networks (CNNs), Deep Neural Networks (DNNs), Recurrent Neural Networks (RNNs), Restricted Boltzmann Machines (RBMs), Deep Belief Networks (DBNs), Bidirectional Recurrent Deep Neural Networks (BRDNNs), or Deep Q-Networks, but are not limited to the examples mentioned above.

[0051] According to an exemplary embodiment of the present disclosure, a processor can implement artificial intelligence. Artificial intelligence refers to a machine learning method based on artificial neural networks that enables a machine to learn by mimicking human biological neurons. Methodologies of artificial intelligence can be classified according to the learning method into supervised learning, where input data and output data are provided together as training data and the solution (output data) to the problem (input data) is predetermined; unsupervised learning, where only input data is provided without output data and the solution (output data) to the problem (input data) is not predetermined; and reinforcement learning, where a reward is given from the external environment whenever an action is taken in the current state, and learning proceeds in a direction that maximizes such reward. Additionally, methodologies of artificial intelligence may be classified according to the architecture, which is the structure of the learning model. The architectures of widely used deep learning technologies can be classified into convolutional neural networks, recurrent neural networks, transformers, generative adversarial networks, etc.

[0052] The device may include an artificial intelligence model. The artificial intelligence model may be a single model or may be implemented as multiple models. The artificial intelligence model may be composed of a neural network (or artificial neural network) and may include statistical learning algorithms that mimic biological neurons in machine learning and cognitive science. A neural network may refer to a model that possesses problem-solving capabilities by having artificial neurons (nodes) that form a network through synaptic connections and change the strength of synaptic connections through learning. The neurons of a neural network may include combinations of weights or biases. A neural network may include one or more layers composed of one or more neurons or nodes. For example, the device may include an input layer, a hidden layer, and an output layer. The neural network constituting the device can infer a result to be predicted from an arbitrary input by changing the weights of the neurons through learning.

[0053] The processor can create a neural network, train or learn a neural network, perform operations based on received input data, generate information signals based on the results of the operations, or retrain the neural network. The neural network models may include, but are not limited to, various types of models such as CNN, R-CNN, RPN, RNN, S-DNN, S-SDNN, Deconvolution Network, DBN, RBM, Fully Convolutional Network, LSTM Network, and Classification Network, such as GoogleNet, AlexNet, and VGG Network. The processor may include one or more processors for performing operations according to the neural network models. For example, the neural network may include a deep neural network.

[0054] Neural networks include CNN, RNN, perceptron, multilayer perceptron, Feed Forward (FF), Radial Basis Network (RBF), Deep Feed Forward (DFF), Long Short Term Memory (LSTM), Gated Recurrent Unit (GRU), Auto Encoder (AE), Variational Auto Encoder (VAE), Denoising Auto Encoder (DAE), Sparse Auto Encoder (SAE), MC (Markov Chain), HN (Hopfield Network), BM (Boltzmann Machine), RBM (Restricted Boltzmann Machine), DBN (Depp Belief Network), DCN (Deep Convolutional Network), DN (Deconvolutional Network), DCIGN (Deep Convolutional Inverse Graphics Network), GAN (Generative Adversarial Network), LSM (Liquid State Machine), ELM (Extreme Learning) Machine), ESN (Echo State Network), DRN (Deep Residual Network), DNC (Differentiable Neural Network) It will be understood by a person skilled in the art that any neural network may be included, but is not limited to, Computer (NTM), Neural Turning Machine (NTM), Capsule Network (CN), Kohonen Network (KN), and Attention Network (AN).

[0055] According to an exemplary embodiment of the present disclosure, the processor comprises a Convolutional Neural Network (CNN) such as GoogleNet, AlexNet, VGG Network, Region with Convolutional Neural Network (R-CNN), Region Proposal Network (RPN), Recurrent Neural Network (RNN), Stacking-based Deep Neural Network (S-DNN), State-Space Dynamic Neural Network (S-SDNN), Deconvolution Network, Deep Belief Network (DBN), Restricted Boltzmann Machine (RBM), Fully Convolutional Network, Long Short-Term Memory (LSTM) Network, Classification Network, Generative Modeling, eXplainable AI, Continual AI, Representation Learning, AI for Material Design, BERT, SP-BERT, MRC / QA, Text Analysis, Dialog System, GPT-3, GPT-4 for Natural Language Processing, Visual Analytics, Visual Understanding, Video Synthesis for Vision Processing, Anomaly Detection, Prediction, Time-Series Forecasting, Optimization, Recommendation for ResNet Data Intelligence, Various artificial intelligence structures and algorithms, such as data creation, may be used, but are not limited thereto. Hereinafter, embodiments of the present disclosure will be described in detail with reference to the attached drawings.

[0056] FIG. 1 is a drawing illustrating that a refund of a deposit is automatically processed according to a recycling processing device (100) according to an embodiment of the present disclosure.

[0057] Referring to FIG. 1, a deposit can be obtained (S14) by purchasing (S11), using (S12), and returning (S13) a recyclable item. For example, the user pays a deposit when purchasing an item that corresponds to a recyclable item, and the deposit is refunded when returning the item after use.

[0058] At this time, the recycling processing device (100) automatically processes the process of returning the recycled material (S13) and refunding the deposit (S14).

[0059] Traditionally, people had to manually check each recyclable item to verify eligibility and process the refund, which required a large workforce and took a significant amount of time.

[0060] However, the recycling processing device (100) according to the embodiment of the present disclosure automatically checks whether the recycled material meets the deposit refund criteria and processes the deposit refund based on the result of the check.

[0061] Below, the operation process of the recycling processing device (100) will be described in more detail with reference to other drawings.

[0062] FIG. 2 is a block diagram of a recycling processing device (100) according to an embodiment of the present disclosure.

[0063] Referring to FIG. 2, a recycling processing device (100) according to an embodiment of the present disclosure includes a processor, a communication module, a memory, a camera, a transfer unit (150), a compression device (160), and a user interface (170).

[0064] However, in some embodiments, the recycling processing device (100) may include fewer or more components than the components shown in FIG. 2.

[0065] The processor (110) controls the entire device and is responsible for major operations for recyclable recognition and classification based on an artificial intelligence model.

[0066] More specifically, the processor (110) receives multiple video frames captured through the camera (140) and converts them into input values ​​for an artificial intelligence model. The processor (110) can perform processes such as classifying recyclable items, classifying detailed items (e.g., beverages / alcohol, domestic / imported, whether a deposit is charged, etc.), detecting contamination, detecting the presence or absence of packaging, classifying colors, and extracting depth information.

[0067] Additionally, the processor (110) can synthesize the inference results to determine whether the recycled material meets the conditions for a refund of the deposit in the country or region.

[0068] For example, the processor (110) can classify items using a YOLOv8-based detection model and then analyze the language / flag / color pattern of the label through OCR and image interpretation to identify whether the product is sold in Canada.

[0069] The processor (110) may be implemented as a memory (130) that stores data for an algorithm or a program that reproduces the algorithm for controlling the operation of components within the device, and at least one processor (110) that performs the aforementioned operation using the data stored in the memory (130). In this case, the memory (130) and the processor (110) may each be implemented as separate chips. Alternatively, the memory (130) and the processor (110) may be implemented as a single chip.

[0070] In addition, the processor (110) may control one or more of the components described above in combination to implement various embodiments according to the present disclosure described in the drawings below on the device.

[0071] In addition to operations related to the application, the processor (110) can generally control the overall operation of the device (100). The processor (110) can provide or process appropriate information or functions to the user by processing signals, data, information, etc. that are input or output through the components described above, or by running an application stored in memory.

[0072] Additionally, the processor (110) can control at least some of the components of the device (100) to run an application stored in memory (130). Furthermore, the processor (110) can operate at least two or more of the components included in the device in combination with each other to run the application.

[0073] The processor (110) may be implemented in one or more ways. Hereinafter, even if the processor (110) is expressed in the singular, it may be considered as plural. The processor (110) can control the configurations of the device (100). The processor (110) may refer to a data processing device embedded in hardware having a physically structured circuit to perform a function expressed by code or instructions included in a program. As such, the processor (110) may encompass processing devices such as a microprocessor, a central processing unit (CPU), a processor core, a multiprocessor, an application-specific integrated circuit (ASIC), and a field programmable gate array (FPGA) as examples of data processing devices embedded in hardware, but the scope of the present invention is not limited thereto. The processor may separately provide a learning processor for performing artificial intelligence computations or may provide a learning processor itself.

[0074] In various embodiments, the processor (110) may include one or more of a Central Processing Unit (CPU), an Application Processor (AP), or a Communication Processor (CP). At least a portion of the processor (110) may be hardware capable of accessing memory (130) and performing functions related to instructions stored in memory (130).

[0075] The communication module (120) can perform the role of transmitting and receiving data to and from an external server or cloud via a wired / wireless network. The communication module (120) can transmit result data collected by the recycling processing device to a server or receive information on changes to the deposit criteria from the server.

[0076] Additionally, the communication module (120) can be linked with an API so that the user can check or view refund information through a mobile device.

[0077] The communication module (120) may include one or more modules that connect the device (100) to one or more networks.

[0078] The communication module (120) may include one or more components that enable communication with an external device, for example, at least one of a broadcast receiving module, a wired communication module, a wireless communication module, a short-range communication module, and a location information module.

[0079] The wired communication module may include various wired communication modules such as a Local Area Network (LAN) module, a Wide Area Network (WAN) module, or a Value Added Network (VAN) module, as well as various cable communication modules such as USB (Universal Serial Bus), HDMI (High Definition Multimedia Interface), DVI (Digital Visual Interface), RS-232 (recommended standard 232), power line communication, or POTS (plain old telephone service).

[0080] In addition to Wi-Fi modules and WiBro (Wireless broadband) modules, the wireless communication module may include wireless communication modules that support various wireless communication methods such as GSM (global System for Mobile Communication), CDMA (Code Division Multiple Access), WCDMA (Wideband Code Division Multiple Access), UMTS (universal mobile telecommunications system), TDMA (Time Division Multiple Access), LTE (Long Term Evolution), 4G, 5G, and 6G.

[0081] The wireless communication module may include a wireless communication interface comprising an antenna and a transmitter that transmit a communication signal. Additionally, the wireless communication module may further include a signal conversion module that modulates a digital control signal output from the processor through the wireless communication interface into an analog wireless signal under the control of the processor.

[0082] A short-range communication module is for short-range communication and can support short-range communication by using at least one of Bluetooth, RFID (Radio Frequency Identification), Infrared Data Association (IrDA), UWB (Ultra-Wideband), ZigBee, NFC (Near Field Communication), Wi-Fi (Wireless-Fidelity), Wi-Fi Direct, and Wireless USB (Wireless Universal Serial Bus) technologies.

[0083] The communication module (120) may include a communication interface.

[0084] A communication interface can establish communication between an electronic device and an external device. For example, the communication interface can communicate with an external device via wireless communication (e.g., Wi-Fi (Wireless Fidelity), Bluetooth, NFC (Near Field Communication), MST (Magnetic Stripe Transmission), etc.) or wired communication.

[0085] The memory (130) is a storage unit that stores artificial intelligence models, judgment logic, setting values, etc. required for the operation of the device, and may include ROM and RAM.

[0086] The memory (130) may store a precision annotation dataset or synthetic data required for artificial intelligence learning / inference.

[0087] Video data collected during operation and processing result logs can be temporarily stored and utilized for user verification or cloud upload.

[0088] The memory (130) stores condition tables corresponding to various country-specific deposit refund standards, and the condition values ​​can be synchronized with an online server.

[0089] The memory (130) can store data that supports various functions of the device. The memory can store a number of application programs (or applications) running on the device, data for the operation of the device, and instructions. At least some of these application programs may exist for the basic functions of the device. Meanwhile, the application programs may be stored in the memory, installed on the device, and driven by the processor to perform operations (or functions).

[0090] The memory (130) can store data supporting various functions of the device and programs for the operation of the processor (110), and can store input / output data (e.g., music files, still images, videos, etc.), and can store a number of application programs (or applications) running on the device (100), data for the operation of the device, and instructions. At least some of these application programs can be downloaded from an external server via wireless communication.

[0091] Such memory (130) may include at least one type of storage medium among flash memory type, hard disk type, SSD type (Solid State Disk type), SSD type (Silicon Disk Drive type), multimedia card micro type, card type memory (e.g., SD or XD memory, etc.), RAM (random access memory; RAM), SRAM (static random access memory), ROM (read-only memory; ROM), EEPROM (electrically erasable programmable read-only memory), PROM (programmable read-only memory), magnetic memory, magnetic disk, and optical disk. Additionally, the memory (130) may be a database that is separated from the device (100) but connected via wired or wireless connection.

[0092] The memory (130) can be electrically connected to the processor (110) and can store at least one code executed by the processor (110). The memory (130) can refer to various types of storage devices. The memory (130) can store information necessary to perform calculations using artificial intelligence, machine learning, or artificial neural networks.

[0093] The memory (130) can store various learning models. The learning models stored in the memory (130) can infer result values ​​for new input data other than the learning data, and the inferred values ​​can be used as a basis for judgment to perform certain actions. Learning can be performed on the learning models stored in the memory (130) based on label information, and various backpropagation algorithms can be applied so that the loss function has a target value in order to increase the accuracy of the learning.

[0094] The camera (140) is an image input means for recognizing the appearance of the recycled material.

[0095] At least one camera (140) may be installed and positioned above the conveyor or at various angles to capture the shape of the recycled material from various angles.

[0096] In an embodiment of the present disclosure, the recycling processing device (100) may include a plurality of cameras or binocular cameras, and depth information (Depth) and 3D shape are extracted and utilized for capacity estimation.

[0097] In some embodiments, an infrared sensor or a near-infrared (NIR) sensor may be used together to perform component determination or transparency analysis.

[0098] Through two RGB cameras installed with top lighting and a ToF sensor-based distance measuring camera, it is possible to determine the aspect ratio, height, and whether it is a curved surface.

[0099] The transfer unit (150) includes a conveyor structure that transfers recycled materials into the device.

[0100] In this embodiment, the conveyor is divided into partitions (compartment structures) at regular intervals so that recyclable items are recognized one by one. This design provides higher recognition accuracy and classification precision than simultaneous processing, and also improves processing speed.

[0101] The transport unit (150) is a timing belt structure and may be transported at a preset speed, or it may be transported after stopping at each section to perform shooting and analysis for a certain period of time.

[0102] The conveying section (150) includes a conveyor belt divided into partitions at regular intervals.

[0103] Multiple cameras (140) are installed on the upper part of the conveyor, and the processor (110) can control the cameras (140) to acquire images and videos of recycled materials from various angles.

[0104] In one embodiment, the processor (110) operates in conjunction with each recognition engine according to the purpose, such as item classification, foreign substance detection, packaging material detection, color classification, and depth measurement, and can comprehensively analyze identification information of recyclables by integrating multiple recognition results.

[0105] The processor (110) can calculate the exact height, volume, etc. of a recycled item through depth measurement. This information can also be used to determine whether the recycled item is deformed, such as whether it is damaged or crushed. For example, even when a PET bottle is crushed and it is difficult to recognize the volume, the processor (110) can infer a relatively accurate circular shape by combining depth information and shape information.

[0106] In addition, the recycling processing device (100) according to the present embodiment can estimate the weight of a recyclable item solely through AI-based inference while continuing to operate the conveyor without a weight detection sensor or a temporary stop operation. For example, a pre-trained AI model can calculate the standard weight or estimated weight of the item based on the shape, size, depth information, etc. of the recyclable item. The weight information can be used to determine refund criteria or to adjust compression pressure.

[0107] The compression device (160) can perform the function of compressing a recyclable material designated as a compression target according to the result of the judgment.

[0108] In particular, the compression device (160) can compress PET bottles, cans, etc. that are larger than a certain size or have a large volume to reduce their volume and optimize storage space.

[0109] At this time, the processor (110) may selectively decide whether to compress based on the result of calculating the weight before compression.

[0110] When the processor (110) determines whether the item is to be compressed, the compression device (160) in the corresponding storage container is activated, and the can is vertically compressed using a piston method.

[0111] The user interface (170) is a display or output device that displays the results of recycling processing, whether the deposit is eligible for refund, the count quantity, etc.

[0112] In some embodiments, the user interface (170) may include a QR issuance or mobile linkage system for refunding the user deposit.

[0113] The user interface (170) visually displays "total estimated refund amount," "list of non-applicable items," "classification success rate," etc., and allows for refund verification through screen touch.

[0114] FIG. 3 is a flowchart of a recycling processing method according to an embodiment of the present disclosure.

[0115] FIGS. 4 to 6 are drawings illustrating the compression of recycled materials transported to a compression device (160) by a transport unit (150).

[0116] FIG. 7 is a drawing illustrating an image of a recycled item being transported by a transport unit (150).

[0117] Figure 8 is a diagram illustrating the final recognition results and refund details for processed recyclables.

[0118] Figure 9 is a diagram illustrating a record of processing through a recycling processing device.

[0119] Referring to FIGS. 3 to 9, a recycling processing device (100) according to an embodiment of the present disclosure will be described.

[0120] The processor (110) controls the transfer unit (150) to transfer the recyclable material. (S100)

[0121] Referring to FIGS. 3 to 6, the conveying unit (150) may include a conveyor (151) partitioned at regular intervals.

[0122] These sections can be formed by partition members (152) having a certain height placed on a conveyor, and each section functions as a unit space for accommodating individual recyclables.

[0123] The processor (110) can use such a partition structure as a reference value to calculate shape information (e.g., height, width, aspect ratio, etc.) of the recyclables. For example, by estimating the relative size of the recyclables in the image based on the fixed width or height of the partition, the correlation with the capacity, volume, or weight of the recyclables can be estimated.

[0124] In addition, the compartment structure prevents the simultaneous recognition of multiple recyclables, enabling accurate unit-by-unit item classification and counting; as a result, improvements in the processing efficiency and recognition accuracy of the entire system can be expected.

[0125] Each section is defined by a partition or physical gap, and is configured so that, in principle, one recyclable item is placed in each section. However, in an actual environment, more than one recyclable item may be located within a section, and in preparation for such a situation, the processor (110) may apply a correction algorithm for removing duplicate recognition or allowing errors.

[0126] The spacing between sections is optimized by considering the camera's shooting range, the resolution of the AI-based recognition engine, and the average size of the recyclables, and the transport speed can also be automatically adjusted to match the recognition and inference processing time.

[0127] The conveyor belt is located inside the device, and a number of cameras and lighting devices can be positioned on the upper side of the belt.

[0128] In addition, since positional variations may occur due to vibration or impact on the belt during transport, an auxiliary processing step that maintains central alignment within the compartment or performs image correction-based position estimation may be applied.

[0129] The processor (110) controls at least one camera to photograph the recyclable material being transported by the transport unit (150). (S200)

[0130] The processor (110) can control at least one camera to acquire a video or image of a recyclable material being transported. At this time, by positioning multiple cameras at different angles on the upper part of the conveyor belt, each camera can photograph the same recyclable material from different viewpoints or directions. For example, one camera can acquire the overall shape of the recyclable material from a vertical direction, while another camera can acquire detailed information such as the shape of the bottleneck and the presence or absence of a lid more precisely from a side or diagonal direction.

[0131] By acquiring video and image information from multiple cameras at various angles in this way, the processor (110) can perform the classification of the types of recyclables and the determination of attributes more accurately using an artificial intelligence model.

[0132] A processor (110) uses an artificial intelligence model to determine the type of recyclable material based on an image captured through a camera. (S300)

[0133] The processor (110) inputs image data obtained through the camera into an artificial intelligence model to classify the types of recyclable materials (e.g., PET bottles, cans, glass bottles, etc.).

[0134] At this time, the processor (110) can distinguish whether it is a beverage can or an aerosol can by analyzing the appearance of a specific can product (label pattern, lid shape, etc.).

[0135] In addition, the accuracy of classification can be improved by additionally considering factors such as shape ratios based on the size of the compartment where each recyclable is located.

[0136] In one embodiment, the processor (110) performs an artificial intelligence-based analysis of the recyclables at step S300, but as a preliminary step for a detailed attribute analysis thereafter, it may only perform a major classification of the types of recyclables.

[0137] For example, the processor (110) identifies whether the recyclable material belongs to one of the major product groups such as cans, glass bottles, PET bottles, plastics, etc., and based on this major classification result, it can be used for processing such as precise classification, identification information analysis, and determination of refund conditions in the subsequent step (S400).

[0138] This major classification method enables improved processing speed and computational resource efficiency of the entire system, and can be particularly advantageous in a structure where multiple classification engines are applied sequentially.

[0139] The processor (110) determines one or more attributes related to the refund criteria for a recyclable item based on the type determined in S300, at least one shape information and at least one identification information for a recyclable item obtainable through an image. (S400)

[0140] The processor (110) can perform detailed classification of recyclables belonging to the corresponding item group in step S400 based on the major classification results performed in step S300.

[0141] For example, if a recyclable item is identified as a can based on the major classification result, the processor (110) can perform detailed item classification such as whether the can is a beverage can, a beer can, or a yogurt can, and further analyze identification information such as metal material (aluminum or iron), manufacturer marking, or distribution information by region (country).

[0142] In an additional embodiment, the processor (110) can accurately determine whether the recyclable belongs to a product family subject to a deposit in a specific country or region by combining visual information (e.g., label, logo, color, etc.), shape information (e.g., capacity estimation), or additional input from a detection sensor (e.g., infrared detection information, etc.) of the recyclable.

[0143] Ultimately, according to the present embodiment, through a sequential structure of major classification of recyclables, detailed item classification, and determination of refund conditions by region, it becomes possible to determine refund conditions flexibly and precisely, capable of responding to various items and national / regional standards.

[0144] In an embodiment of the present disclosure, the shape information may include at least one of the outer contour, aspect ratio, height, width, or whether the surface is curved of the recycled product.

[0145] The processor (110) can identify the overall shape of the object, whether it is a closed curve, curvature, step difference, etc., by extracting the outline of the recycled material within the image frame. This is useful for classifying differences in shape, such as glass bottles, cans, PET, etc.

[0146] The processor (110) can calculate the ratio of the length to the width of the recycled material to classify whether it is a bottle shape (long vertical structure) or a cylinder shape (short horizontal and relatively low height structure).

[0147] The processor (110) aligns image information acquired by multiple cameras at different locations, and the processor (110) can estimate the height or floor area of ​​the recycled material.

[0148] In particular, for objects with a consistent shape, such as PET bottles and aluminum cans, the height can be inferred from the base area alone, which also allows for volume estimation.

[0149] The processor (110) can determine whether a rounded surface exists through a contour or three-dimensional structure recognition algorithm and can be used to distinguish it from non-recyclable items such as flat packaging paper.

[0150] In an embodiment of the present disclosure, the identification information may include at least one of a label, character, barcode, logo, or color information included in the recycled item.

[0151] According to one embodiment, the processor (110) analyzes at least one of a label, character, barcode, logo, or color information included on the surface of a recyclable item to obtain identification information for the recyclable item in step S400.

[0152] For example, when a recyclable item moving along a conveyor belt (130) is captured by at least one camera (120), the processor (110) can extract character and color features from the captured image to identify whether the recyclable item is a product of a specific brand.

[0153] This identification information is useful for distinguishing the manufacturer, country of sale, product type, and deposit eligibility depending on the product.

[0154] In addition, some cans or bottles may contain identification information such as barcodes or QR codes, and the processor (110) detects the barcode area in the image and interprets the actual value using an optical character recognition (OCR) or barcode decoding algorithm, thereby more accurately determining whether the recyclable item is a domestically distributed product or a foreignly distributed product, or whether it is a product with a deposit.

[0155] For example, even if the shape and material of the cola can are the same, if the language printed on the surface is English and the barcode is confirmed to be a standard for the North American region, the processor (110) can determine that the product is imported from Canada or the United States and identify that it is not eligible for a refund of the deposit.

[0156] As another example, if a specific colored band (e.g., a blue band) is displayed on the surface of a bottle, this can be used as a color recognition standard to identify products registered in a local deposit system, and the processor (110) can detect the color information to more efficiently determine whether the deposit refund conditions are met.

[0157] In this way, the present embodiment acquires identification information through image-based recognition, and can precisely select products eligible for refund by region by combining various visual elements.

[0158] According to one embodiment, the processor (110) determines the major classification type of the recyclable material in step S300, and then in step S400, can perform a more precise determination by referring to prior shape information corresponding to the type.

[0159] For example, if the processor (110) determines in step S300 that a specific recyclable is a PET bottle, the PET bottle generally has a cylindrical shape, and the ratio of the diameter of the opening to the body, the height range, etc., may be stored as pre-registered template information. This information is stored in memory (140) and is automatically recalled in conjunction with the type determination result.

[0160] In this case, the processor (110) can inversely estimate the height or volume of the PET bottle by utilizing the width (diameter) information of the cross-section obtained within the PET bottle image and the pre-registered ratio information. For example, if the width is measured as 6 cm, the height is estimated to be about 18 cm by using the pre-registered average ratio of the PET bottle (height:width = 3:1), which can then lead to the calculation of the volume.

[0161] Likewise, in the case of a can, the processor (110) can infer the total height and capacity by obtaining only the cross-sectional diameter based on the typical cylindrical shape of the can and common specifications (diameter, height, etc.) of each manufacturer. This method enables weight and volume estimation with significantly high accuracy using only image-based inference, without the need for an actual weight sensor.

[0162] In this way, the "attribute determination" in step S400 does not rely solely on shape information obtained from the current image, but derives extended information through comparison or correction with a corresponding pre-template or reference value depending on the type of recycled material, and this information serves as a decisive judgment criterion in determining the deposit refund condition (S500).

[0163] Furthermore, this inference can be performed more precisely by combining image frames from various angles (front, side, oblique, etc.) provided by multiple cameras, and, for example, height can be obtained more accurately from the side camera, while diameter or surface labels can be obtained more accurately from the front camera.

[0164] In one embodiment, the processor (110) can determine attributes necessary for determining the deposit refund condition by obtaining first shape information of a recyclable item through image information and inferring complementary second shape information based on the first shape information by referring to reference shape information corresponding to the type of recyclable item determined.

[0165] For example, the processor (110) can combine the acquired first shape information (e.g., diameter of a can) with reference shape information and, if the product is similar to the reference shape, can complementarily infer second shape information (e.g., height or volume). For example, if the diameter is measured to be 6.5 cm and the average diameter of that type is 6.5 cm, the reference height of 12 cm can be inferred as second shape information.

[0166] The second shape information inferred in this way (e.g., height or total volume) can then be compared with the standard values ​​of the deposit refund conditions (e.g., 355 ml or more, 500 ml or more, etc.) to determine whether the recyclable item is eligible for a refund.

[0167] Through this, the processor (110) can perform a more accurate determination of refund conditions by combining image-based first shape information and prior reference data without using complex sensors separately.

[0168] For example, the attribute is a value related to the capacity or weight of the recyclable, and the processor (110) can determine whether the refund condition is met based on whether the attribute satisfies the value defined in the recyclable refund criteria.

[0169] The processor (110) determines whether the attribute determined in S400 satisfies the deposit refund conditions of the recyclable item. (S500)

[0170] According to one embodiment, the processor (110) can determine whether the recyclable item satisfies the deposit refund conditions based on the attributes of the recyclable item determined in step S400 (e.g., type, volume, weight, label information, etc.).

[0171] For example, if the type of recyclable inferred from S400 is an aluminum can, the estimated capacity is greater than or equal to the capacity set as the deposit refund standard, and a regional code (e.g., "CAN" or "BC") is detected through the character / color information on the surface, the processor (110) can determine that it is a 'deposit-eligible can sold in Canada' and determine that the deposit refund conditions are met.

[0172] On the other hand, even if a recycled item has the same shape, if "USA" or "IMPORT" is detected in the barcode or text information on the surface, it is determined that the deposit refund conditions are not met. In this way, the processor (110) can examine whether the conditions are met from various angles based not only on the shape but also on a combination of multiple attribute information.

[0173] For example, the processor (110) can calculate at least one of the capacity or weight of a recycled item from shape information and determine whether the recycled item meets the deposit refund conditions based on at least one of the calculated capacity or weight.

[0174] The processor (110) analyzes image information obtained from the camera to extract shape information of the recycled material. For example, it can calculate the approximate capacity by measuring the bottom diameter and height of a PET bottle, or estimate the weight based on the overall size and shape of the object.

[0175] The estimated capacity or weight is compared with the deposit refund criteria to determine whether the recyclable item is eligible for a refund. For example, if the capacity is estimated to be 500ml or more, it is eligible for a refund; otherwise, it is excluded.

[0176] As a result of the processor (110) performing the process of S500, if the recyclable item satisfies the recyclable item refund conditions, the refund amount is counted. (S600)

[0177] When the processor (110) determines in S500 that the recyclable item meets the deposit refund conditions, it counts the cumulative refund count. For example, whenever a recyclable item eligible for refund is processed, it can be recorded in the counter of the user account or the current equipment session.

[0178] Additionally, if the refund amount may differ depending on the type of recycled material, the processor (110) can calculate the total accumulated refund amount by classifying the amounts based on the information. For example, aluminum cans are counted as 10 cents and plastic bottles as 20 cents.

[0179] The counting result is displayed to the user in real time through the user interface (170) and can be converted into points that the user can receive as a refund through a QR code, app integration, or reward system.

[0180] In one embodiment, the processor (110) can determine whether a product is eligible for regional refund based on at least one of a label, character, logo, or color information included on the surface of a recycled product.

[0181] The processor (110) analyzes labels, text, logos, or color information attached to the surface of a recyclable item in an image acquired through a camera. For example, it can determine whether text, a brand logo, or a specific color pattern printed on a PET bottle is included.

[0182] This information is compared against a pre-registered list of regional deposit-refund eligible products. For example, if phrases such as 'CANADA REFUND' are recognized or a specific logo (e.g., a Canadian government approval mark) is identified, the recyclable item is determined to be eligible for a deposit refund in Canada.

[0183] Conversely, if the processor (110) does not have the corresponding label or text, or if the color or design does not match the refundable product family, the processor (110) may determine that the recyclable item is excluded from refund.

[0184] In addition, deposit refund criteria may include whether the product falls under a product group subject to a deposit at the time of sale in the relevant region or country.

[0185] This approach is effective in allowing for flexible responses to refund policies that vary by country or region.

[0186] A recycling processing device (100) according to an embodiment of the present disclosure can store image and result data in a log format during the recycling recognition process in order to effectively respond to a client's complaint situation.

[0187] Specifically, the processor (110) can control the camera (140) to acquire multiple images during the process of the recycled material moving through the transport unit (150).

[0188] Referring to FIG. 7, an example is shown of a processor (110) capturing an image of a recyclable moving on a conveyor through a camera (140). At this time, objects such as cans, PET bottles, and bottles may be labeled with a detection box, and each object may have a recognized class or unique ID and sequence number displayed.

[0189] The processor (110) inputs multiple acquired images into an artificial intelligence-based recognition model, and the recognition model can produce independent recognition results for each image.

[0190] Then, the processor (110) aggregates the multiple recognition results to determine the final recognition result. At this time, a voting algorithm may be used as the determination method, and if multiple identical recognition results are obtained among multiple images, this is determined as the final recognition result.

[0191] For example, if five images are acquired for a specific recyclable item and a recognition result is derived for each image, if the same result is derived from three or more images, the result is confirmed as the final recognition result. Subsequently, the processor (110) may store at least one of the images that produced a recognition result matching the final recognition result as a log image in memory (130).

[0192] The log image stored in the memory (130) can subsequently be used as evidence to prove that the recyclable material was correctly classified and determined when responding to customer complaints. Additionally, the processor (110) may, if necessary, select the image with the highest recognition accuracy among multiple images as the log image.

[0193] In one embodiment, the processor (110) can store the details of processing through the recycling processing device (100) and generate statistical data.

[0194] By storing the log image in this manner, the recycling processing device (100) according to the present embodiment can go beyond recognizing and classifying recyclables to ensure transparency of the results and the possibility of post-verification.

[0195] In one embodiment, the processor (110) may store the recyclable item recognition image or video obtained through the camera (140) as is. The video stored at this time may include a plurality of recyclable items moving along the transport unit (150), and the processor (110) may manage the recyclable item-specific sections included in the video by matching them with the recognition results.

[0196] Through this, when the recycling processing device (100) needs to verify a record of a specific recyclable, it can identify the time and section where the recyclable appears in the memory (130) and verify the video or image of that section. According to this embodiment, the reliability of the recyclable recognition result can be increased, and it can be used as supporting evidence in the event of a client's request for verification or a dispute.

[0197] Additionally, the recycling processing device (100) can provide information on the final recognition result and refund details for the recycled material processed as in FIG. 8, and can record the details processed through the recycling processing device (100) as in FIG. 9.

[0198] Referring to FIG. 8, the processor (110) can input multiple video frames captured through the camera (140) into an artificial intelligence model to perform item classification for recyclables and display the result on the user interface (170).

[0199] In detail, the processor (110) can display each recyclable item, such as aluminum, beer can, plastic, and glass, in a list format, and display the quantity (Qty) and amount (Amount) for each item together so that the refund amount can be checked immediately.

[0200] The processor (110) can display the accuracy of recognition during the process of handling the input recyclables.

[0201] The processor (110) can count and display the cumulative number of items of recyclables.

[0202] Referring to FIG. 9, information that allows for managing, verifying, and analyzing the history of previously recognized recyclable materials is displayed through the user interface (170), and the total amount (Total) and user (User) are displayed by date, and one of a plurality of actions can be selected together.

[0203] At this time, when the playback function is executed, the processor (110) can display the stored recycled item image for the case.

[0204] Referring to the right image of FIG. 9, the processor (110) can generate statistics by type of processed recyclables and can also display statistical information on whether refund conditions are met.

[0205] In one embodiment, the processor (110) can identify the type of recyclable material by analyzing an image captured through a camera (140) using an artificial intelligence model. The processor (110) can accumulate and manage the number of recyclable materials by type identified in this way.

[0206] Specifically, whenever each recyclable item transported through the transport unit (150) is recognized, a count value corresponding to that type increases, and the processor (110) stores the accumulated count value in memory (120) in real time. Additionally, the accumulated count of each type up to now can be visually displayed through the user interface (160).

[0207] In one embodiment, the processor (110) can classify each recyclable material fed through the transfer unit (150) by analyzing it with an artificial intelligence model. The classified results are displayed in real time through the user interface (160), and the user can refer to them to classify the recyclable materials into corresponding bags.

[0208] Additionally, the processor (110) can control the increase in the count corresponding to the type whenever each recyclable is identified as a specific type, and the increased count value to be immediately reflected in the user interface (160). At this time, the user interface (160) can be configured to visually display the count value by type so that the number of recyclables classified so far in each bag and the target quantity can be checked simultaneously.

[0209] Therefore, the user can check the number of recyclables placed in the bag without any separate manual work, and can easily determine whether the set quantity has been met. According to this embodiment, the efficiency of the recyclable sorting and packaging process can be improved, and the risk of penalties due to quantity discrepancies can be minimized.

[0210] For example, the user can put each recyclable item into the corresponding bag from behind the recognition device while checking the cumulative count by type displayed on the user interface (160). Each bag has a predetermined count standard, and the user can easily determine whether the number of recyclable items classified in the bag has reached the standard count by checking the cumulative count.

[0211] According to this embodiment, even if random inspections are conducted when recyclable materials are packaged in bags and transported to a center, count errors can be minimized based on the cumulative count data provided by the recognition device. As a result, recognition accuracy and data-based management become possible, thereby reducing the risk of penalties caused by incorrect counts.

[0212] The processor (110) controls the transfer unit (150) to transfer the recyclable material to the compression device (160) and controls the compression device (160) to compress the recyclable material. (S700)

[0213] According to one embodiment, the processor (110) controls the transfer unit (150) to transfer the recyclable material toward the inlet of the compression device (160). At this time, all recyclable materials may undergo a compression process regardless of whether a refund is determined, or, if necessary, recyclable materials that are not eligible for a refund may be configured to be discharged via a separate path.

[0214] When the transfer is complete, the processor (110) operates the compression device (160) to compress the recycled material. The compression device (160) may include, for example, a vertical or horizontal hydraulic press and may automatically adjust the compression strength according to various materials (e.g., plastic, metal).

[0215] After compression, the recyclables are moved to a designated bin, and at this time, the completion of compression can be reconfirmed via a camera or sensor. The completion status is logged internally or transmitted to the operator to support the management of the equipment's processing history.

[0216] According to one embodiment, the transfer unit (150) may be structured to include a plurality of storage containers and compression devices (160). Each storage container or compression device (160) is arranged separately according to the type of recyclable material and whether the refund conditions are met.

[0217] The processor (110) determines the type of recyclable material at step S300 and evaluates whether the recyclable material satisfies the deposit refund conditions at step S500, and then controls the transfer unit (150) in the following manner.

[0218] ① Branching treatment according to the type of recyclable material

[0219] For example, if the processor (110) determines that the recyclable item is an aluminum can or a PET bottle, since the item is a target for volume reduction through future compression processing, it controls the transfer unit (150) to transfer the recyclable item to the compression device (160).

[0220] At this time, the compression device (160) may be divided according to the material and divided into a device for cans / a device for plastics.

[0221] On the other hand, items that are inefficient to compress, such as glass bottles, are not sent to the compression device (160), but are directly transferred to the storage container by controlling the transfer unit (150).

[0222] ② Branching based on whether refund conditions are met

[0223] If the deposit refund conditions are met, the processor (110) branches the recyclable material to a refund target storage path.

[0224] Example: PET bottles that meet the conditions are transferred to a compression device (160) and compressed

[0225] Example: Glass bottles that meet the conditions are transferred to the refund eligible storage box.

[0226] Conversely, if the recycled material does not meet the deposit refund conditions, the processor (110) can control the transfer unit (150) to process it in the following manner.

[0227] (a) Transfer to a dedicated bin for non-refundable items to guide the user to retrieve them, or process for general recycling.

[0228] (b) Compression or storage will proceed as planned, but will be excluded from the refund count. In other words, pure collection will be performed, but no compensation will be paid.

[0229] For example, when a PET bottle is photographed through a camera and the processor (110) recognizes a Canadian deposit system target label ("BC", "CAN", "Deposit", etc.), if it is determined that the refund conditions are met, the PET bottle is transferred to the corresponding compression device (160) and a refund of 0.2 dollars is counted.

[0230] For example, if the same PET bottle is detected but has a label such as "USA" or "No Deposit", the processor (110) determines that the item is not eligible for a refund and transfers it to a general PET collection sorting bin or to the same compression device (160) without counting it for a refund.

[0231] For example, in the case of bottled products, refund eligibility is determined through regional code and shape analysis, and bottles that meet the criteria are transferred to a storage locker. Bottles that do not meet the criteria are transferred to a separate sorting bin or a general storage route.

[0232] Through the above embodiment, the processor (110) can intelligently adjust the transport path by considering not only the classification of the types of recyclables but also whether the refund conditions are met. As a result, the user receives an accurate deposit refund service, and the system operator can secure both compression efficiency and classification accuracy. This is one of the core structures that constitute the technical differentiation of the automatic recyclable sorting and reward system.

[0233] The method according to one embodiment of the present disclosure described above may be implemented as a program (or application) and stored on a medium to be executed in combination with a server, which is hardware.

[0234] The aforementioned program may include code encoded in a computer language such as C, C++, JAVA, or machine language, which can be read by the computer's processor (CPU) through the computer's device interface, in order for the computer to read the program and execute the methods implemented in the program. Such code may include functional code related to functions that define the necessary functions for executing the methods, and may include control code related to execution procedures necessary for the computer's processor to execute the functions according to a predetermined procedure. Additionally, such code may further include memory reference code regarding where (address) additional information or media necessary for the computer's processor to execute the functions should be referenced in the computer's internal or external memory. In addition, if the processor of the computer needs to communicate with any other computer or server located remotely in order to execute the above functions, the code may further include communication-related code regarding how to communicate with any other computer or server located remotely using the communication module of the computer, and what information or media to transmit or receive during communication.

[0235] The above-mentioned storage medium refers to a medium that stores data semi-permanently and is readable by a device, rather than a medium that stores data for a short period of time, such as a register, cache, or memory. Specifically, examples of the above-mentioned storage medium include, but are not limited to, ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage device. That is, the above-mentioned program may be stored on various recording media on various servers that the computer can access, or on various recording media on the user's computer. Additionally, the above-mentioned medium may be distributed across networked computer systems, and computer-readable code may be stored in a distributed manner.

[0236] The steps of the method or algorithm described in connection with the embodiments of the present disclosure may be implemented directly in hardware, implemented as a software module executed by hardware, or implemented by a combination thereof. The software module may reside in RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), Flash Memory, a hard disk, a removable disk, a CD-ROM, or any form of computer-readable recording medium well known in the art to which the present disclosure belongs.

[0237] Although embodiments of the present disclosure have been described above with reference to the attached drawings, those skilled in the art will understand that the present disclosure may be implemented in other specific forms without altering its technical concept or essential features. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive.

Claims

1. A conveying unit for conveying recyclable materials; At least one camera for photographing recyclables transported by the above-mentioned transport unit; and Using an artificial intelligence model, the type of the recyclable material is determined based on the video captured by the camera, and Based on the determined type, at least one shape information and at least one identification information regarding the recyclable item obtainable through the image, one or more attributes related to the deposit refund criteria of the recyclable item are determined, and A processor including a processor that determines whether the above-determined attribute satisfies the above-determined deposit refund conditions, Recycling processing device.

2. In Paragraph 1, The above identification information is, at least one of a label, character, barcode, logo, or color information included in the above-mentioned recyclable material, Recycling processing device.

3. In Paragraph 1, The above processor is, Acquire first shape information of the recycled product through the above image information, and Characterized by determining the attributes necessary for determining the deposit refund conditions by inferring complementary second shape information based on the first shape information by referring to standard shape information corresponding to the type of recyclable material determined above. Recycling processing device.

4. In Paragraph 1, The above shape information is, Characterized by including at least one of the outer contour, aspect ratio, height, width, or whether the surface of the above-mentioned recycled material is curved. Recycling processing device.

5. In Paragraph 1, The above processor is, Calculate at least one of the capacity or weight of the recycled material from the shape information above, and Characterized by determining whether the deposit refund conditions are satisfied based on at least one of the above-calculated capacity and weight, Recycling processing device.

6. In Paragraph 1, The above processor is, Characterized by determining whether a product is eligible for regional refund based on at least one of a label, character, logo, or color information included on the surface of the above-mentioned recyclable product. Recycling processing device.

7. In Paragraph 1, The above transfer unit is, It includes a conveyor divided into sections at regular intervals, The above processor is, Characterized by using the above-mentioned compartment structure as a reference value for determining the above-mentioned shape information, Recycling processing device.

8. In Paragraph 1, The above deposit refund criteria are, Including whether it falls under the product family subject to a deposit at the time of sale in the relevant region or country, Recycling processing device.

9. In Paragraph 1, The above attributes are, It is a value related to the capacity or weight of the above-mentioned recyclable material, and The above processor is, Characterized by determining whether the refund condition is satisfied based on whether the above attribute satisfies the value defined in the above deposit refund criteria, Recycling processing device.

10. A method performed by a recycling processing device, A step of transporting recyclables by controlling a transport unit; A step of controlling at least one camera to photograph the recyclable material being transported by the transport unit; A step of determining the type of recyclable material based on an image captured through the camera using an artificial intelligence model; A step of determining attributes related to the deposit refund criteria of the recyclable item based on at least one shape information of the recyclable item obtainable through the determined type and the image; and A step including determining whether the above-determined attribute meets the above-determined deposit refund conditions, Control method for a recycling processing device.

Images